Synergistic Application of XPS and DFT to Investigate Metal Oxide Surface Catalysis

To investigate metal oxide surface catalysis, determining an appropriate Hubbard U-correction term is a challenge for the density functional theory (DFT) community and identifying realistic reaction intermediates and their corresponding X-ray photoelectron spectroscopy (XPS) shifts is a challenge for experimental researchers, when these methods are used independently. In this study, using CuO as a model transition metal oxide, we demonstrate that when DFT and XPS are applied synergistically, the determination of the U value and the identification of adsorbate/intermediate species on the surface (and their XPS shifts) can be done simultaneously. The experimental O 1s spectra of the as-synthesized CuO 2D-nanoleaves shows the presence of four different peaks with core level binding energies (CLBEs) of 529.7, 531.4, 533.2, and 534.6 eV. DFT is used to calculate the CLBE shifts for probable adsorbed moieties, in various adsorption configurations, on both, clean and vacancy defect containing surfaces. Comparison of experimental and theoretical CLBEs across the entire U value range of 0-9 eV narrows down the list to only four moieties, namely, O-2 in the eta(1)(O) configuration, H2O at the surface oxygen vacancy site, and adsorbed HCO3 and HCO2 (resembling adsorbed HCO3). Finally, the U value of 4-4.5 eV reproduces the experimental CLBE shifts correctly and thus, establishes these experimental XPS spectral peaks to the adsorbates and their geometries. The integrated approach elucidated in this article, results in the identification of adsorbates/intermediates (and their CLBEs) for the experimental XPS spectral analysis and the determination of an appropriate U value concurrently, to study metal oxide surface catalysis.